Gaze Defect Compensation via Dynamic Eye Model Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Eye tracking systems face performance issues when dealing with users who have health-related eye defects such as Strabismus, as existing methods struggle to accurately estimate gaze direction due to anatomical variations and imperfections in the human eye.
Innovation Solution
The implementation of a gaze-dependent eye-model that selects and applies different calculation processes based on the direction of the optical axis, using a plurality of eye models and gaze offset values to compensate for Strabismus and other gaze-related defects by determining and applying appropriate eye modeling parameters during a calibration process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single eye tracking model is used for all users, then the system structure remains simple, but measurement precision deteriorates for users with eye defects such as Strabismus
Solution Approach 1:
The patent divides the eye tracking model into multiple segmented models, each optimized for specific gaze directions or eye conditions. Instead of using a single universal model, the system segments the gaze space into multiple regions and assigns different models to each region, thereby improving measurement precision for users with eye defects while managing complexity through modular organization.
Solution Approach 2:
The patent implements a dynamic model selection mechanism that adapts the eye tracking model based on the detected gaze direction. The system dynamically selects or switches between different eye models depending on the current gaze region, allowing the measurement precision to be optimized for each specific gaze direction rather than using a static single model for all conditions.
2Measurement precision
If multiple eye models are implemented to compensate for eye defects, then measurement precision improves for users with Strabismus, but device complexity increases
Solution Approach 1:
The patent applies local quality by using different eye models for different gaze directions or eye conditions. Each local region of the gaze space has its own optimized model, allowing the system to achieve high measurement precision for users with eye defects in their specific gaze regions without requiring all possible models to be active simultaneously, thus managing complexity.
Solution Approach 2:
The patent changes parameters of the eye tracking model based on detected eye characteristics during calibration. By adjusting model parameters according to the user's specific eye geometry and gaze patterns, the system achieves high precision for users with eye defects without necessarily requiring multiple complete models, thereby controlling complexity through parameter adaptation rather than model proliferation.
3Adaptability or versatility
If gaze direction-dependent model selection is implemented, then adaptability improves for different gaze directions, but computational complexity increases
Solution Approach 1:
The patent performs preliminary actions by pre-calculating and storing multiple eye models for different gaze directions during the calibration phase. During actual operation, the system only needs to select from these pre-prepared models based on the current gaze direction, which is determined through relatively simple geometric calculations. This preliminary preparation reduces the computational complexity during real-time tracking while maintaining high adaptability.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
An eye tracking system comprising a controller configured to, for one or both of a left eye and a right eye of a user: receive eye measurement data associated with the eye; determine an optical axis of the eye from the eye measurement data; and select one of a plurality of eye models based on a direction of the optical axis; and determine a gaze vector of the eye by applying the selected eye model to the eye measurement data.